Skip to content

Authoring class

Writes exactly one artifact into the repo, then proposes the commit — never runs git itself. Shared rules: CONVENTIONS-authoring.md.

10 skills

ingest→draft→propose

What it does

Writes an Agentic design section (loop bounds, tool contracts, state/graph, eval plan, harness) into the spec or plan it is given, or into docs/design/agentic-design.md, and proposes…

Agents

runs inline

— · v0.1.0
✍️

/arch-diagram

ingest→draft→propose

What it does

Reads the project's README, architecture docs, and ADRs first to ground the diagram, then authors a presentation-quality SVG directly — infographic style by default, --sketch for a…

Agents

diagram-architect

0–1 agent calls — the diagram is normally authored inline, and `diagram-architect` is spawned only for a large or unfamiliar system · v0.4.0
ingest→draft→propose

What it does

Captures from a running web app, terminal/CLI output, a static HTML/component, or existing images, then runs a redaction pipeline (detect PII/secrets → blur/mask → verify) so nothing…

Agents

screenshot-capturer · screenshot-redactor · sensitive-data-reviewer

a 3-agent pipeline (capturer → sensitive-data-reviewer → redactor) per image · v0.1.0
✍️

/changelog

ingest→draft→propose

What it does

Generates or updates CHANGELOG.md using the Keep a Changelog format + Semantic Versioning, reading Conventional Commits as the input signal, then PROPOSES the commit (never commits…

Agents

changelog-writer

1–2 agent calls · v0.1.0
✍️

/docs-site

ingest→draft→propose

What it does

By DEFAULT it builds a generated multi-page site (MkDocs Material + gen-files + literate-nav) with a tree/sidebar nav, one page per entity, and embedded diagrams — rebuilt from the…

Agents

docs-architect · docs-designer

1–2 agent calls · v0.3.0
ingest→draft→propose

What it does

Produces a verified, sourced, illustrated weekly intelligence report for an Engagement Manager covering their book of accounts, a market pulse, leadership moves and the competitive…

Agents

runs inline

— · v1.0.0
ingest→draft→propose

What it does

Grounds the security/governance layer in the OpenSSF OSPS Baseline (cited by control ID, Level 1 by default) and delegates the stack-specific layout to the ecosystem's own generator…

Agents

runs inline

— · v0.1.0
ingest→draft→propose

What it does

Detects what changed since the docs were last updated, captures fresh screenshots of the affected UI with a headless browser, and edits the relevant doc sections in place.

Agents

runs inline

— · v0.1.0
✍️

/tech-blog

ingest→draft→propose

What it does

Runs a multi-agent pipeline (writer → fact-checker → reviewer → editor → optional poster), grounds every claim in the actual repo (the fact-checker BLOCKS on anything it can't verify)…

Agents

blog-editor · blog-fact-checker · blog-final-polish · blog-platform-lint · blog-poster · blog-reviewer · blog-writer · diagram-architect · social-post

a 6-agent pipeline (writer → fact-checker → reviewer → editor → final-polish → platform-lint), plus one extra writer+fact-checker round per fact-check retry (capped at 2, `§C`) · v0.2.0
ingest→draft→propose

What it does

Use this skill when the user asks for \"vertical pulse FinTech\", \"vertical pulse for Travel & Hospitality\", \"this week in \", \"what moved in this week\", or…

Agents

runs inline

— · v1.0.0